Comparative analysis of the peanut witches'-broom phytoplasma genome reveals horizontal transfer of potential mobile units and effectors

Abstract

Phytoplasmas are a group of bacteria that are associated with hundreds of plant diseases. Due to their economical importance and the difficulties involved in the experimental study of these obligate pathogens, genome sequencing and comparative analysis have been utilized as powerful tools to understand phytoplasma biology. To date four complete phytoplasma genome sequences have been published. However, these four strains represent limited phylogenetic diversity. In this study, we report the shotgun sequencing and evolutionary analysis of a peanut witches'-broom (PnWB) phytoplasma genome. The availability of this genome provides the first representative of the 16SrII group and substantially improves the taxon sampling to investigate genome evolution. The draft genome assembly contains 13 chromosomal contigs with a total size of 562,473 bp, covering ∼90% of the chromosome. Additionally, a complete plasmid sequence is included. Comparisons among the five available phytoplasma genomes reveal the differentiations in gene content and metabolic capacity. Notably, phylogenetic inferences of the potential mobile units (PMUs) in these genomes indicate that horizontal transfer may have occurred between divergent phytoplasma lineages. Because many effectors are associated with PMUs, the horizontal transfer of these transposon-like elements can contribute to the adaptation and diversification of these pathogens. In summary, the findings from this study highlight the importance of improving taxon sampling when investigating genome evolution. Moreover, the currently available sequences are inadequate to fully characterize the pan-genome of phytoplasmas. Future genome sequencing efforts to expand phylogenetic diversity are essential in improving our understanding of phytoplasma evolution.

Figure 1. Molecular phylogeny of phytoplasmas.

The maximum likelihood tree is inferred based on the 16S ribosomal RNA gene. GenBank accession numbers are listed in square brackets following the species names. The numbers on the internal branches indicate the percentage of bootstrap support based on 10,000 re-sampling (only values >80% are shown). The sequence from Acholeplasma laidlawii is included as the outgroup. The species with genome sequences available are highlighted in bold (*partial; **complete). Abbreviations: peanut witches'-broom phytoplasma strain NTU2011 (PnWB NTU2011); sweet potato little leaf phytoplasma strain V4 (SPLL V4); milkweed yellows phytoplasma strain MW1 (MY MW1); Vaccinium witches'-broom phytoplasma strain VAC (VacWB VAC); Italian clover phyllody phytoplasma strain MA1 (ICP MA1); poinsettia branch-inducing phytoplasma strain JR1 (PoiBI JR1).

Figure 2. Functional classification of annotated protein-coding genes according to COG assignments.

Genes without inferred COG annotation were assigned to a custom category X. The number of protein-coding genes in each set is labeled in the center of each pie chart. (A) All 421 annotated protein-coding genes in the PnWB NTU2011 genome. (B) The 226 protein-coding genes that have specific functional category assignments.

Figure 3. Distribution patterns of homologous gene clusters.

Panels (A) and (B) are based on all protein-coding genes, panels (C) and (D) are based on the putative effectors. Panels (A) and (C) illustrate the phylogenetic distribution patterns of homologous gene clusters. The numbers above a branch and preceded by a ‘+’ sign indicate the number of clusters that are uniquely present in all daughter lineages, the numbers below a branch and preceded by a ‘−’ sign indicate the number of clusters that are uniquely absent. For example, in panel (A) PnWB NTU2011 and ‘Ca. P. mali’ share six homologous gene clusters that are not found in any of the other three phytoplasma genomes; similarly, 36 clusters are absent in both PnWB NTU2011 and ‘Ca. P. mali’ but are shared by the other three phytoplasma genomes. Panels (B) and (D) illustrate the number of homologous gene clusters shared between each genome-pair, numbers on the diagonal indicate the number of clusters found in each individual genome.

Figure 4. Highlights of selected metabolic pathways and transporters in phytoplasmas.

The color-coded circles indicate the presence (filled) or the absence (unfilled) of a gene in each genome. Genes that are present in the PnWB phytoplasma and absent in the other four genomes are highlighted in red.

Figure 5. Molecular phylogeny of conserved and PMU-associated genes.

The numbers on the internal branches indicate the percentage of bootstrap support based on 1,000 re-sampling (only values >80% are shown). (A) The organismal phylogeny based on the concatenated alignment of 214 single-copy genes conserved among the five phytoplasma genomes and the outgroup A. laidlawii (75,565 aligned amino acid sites). The tree topology is consistent with that inferred from 16S rRNA genes (Figure 1). (B) Molecular phylogeny of PMU-associated ATP-dependent Zn protease (hflB; 855 aligned amino acid sites). (C) Molecular phylogeny of PMU-associated replicative DNA helicase (dnaB; 533 aligned amino acid sites). (D) Molecular phylogeny of PMU-associated DNA primase (dnaG; 614 aligned amino acid sites). Note that the outgroup is not included in panels (B)-(D) because no PMU was found in the A. laidlawii genome.